Life-threatening infections and sepsis take an enormous toll on Americans each year. Although often considered a disease of intensive care units, population-based estimates suggest as many as 500,000 septic patients are cared for in emergency departments annually. A cornerstone of sepsis is widespread dysregulation of pro- and anti-inflammatory mechanisms. Among these, the complement cascade is of significant interest, largely because of the multitude of preclinical therapies directed against its role in sepsis. In this first competing renewal, our overall goal remains to better understand the complement cascade, and specifically the activation product C5a, in acute life-threatening infection and sepsis. Complement is crucial for early bacterial detection, host signaling, and pathogen eradication. However, extensive evidence exists in rodent models (and to a much lesser degree in septic patients) that its dysregulation can lead to pathologic humoral and cellular effects which may actually increase the lethality of systemic infection. Novel therapies targeting both C5a and its receptors are in development, but clinical understanding of complement activation in this population is surprisingly limited. Our focus in this renewal is severely septic patients being evaluated in the emergency department. We specifically propose three aims. First, we intend to enroll 150 patients with severe sepsis and 150 patients with non-septic illness and to identify clinical and genetic risk factors for significant C5a production and to correlate these features with clinical course. This aim will include parallel measurement of blood neutrophil C5a receptor expression. In the second aim, we will examine the features of bacterial surfaces that contribute to or inhibit the activation of complement and the production of C5a during infection. The work will include novel assays of host bactericidal activity and C5a generation and will measure function in a subset of patients from our first aim. In our third aim, we will approach the problem of C5a generation on bacterial surfaces from an altogether different perspective, by examining for the first time crosstalk between the adrenomedullin pathway and the alternative complement pathway, which are linked by the complement regulatory protein Factor H, which has recently been identified as an adrenomedullin binding protein necessary for the full vascular smooth muscle effects of adrenomedullin. The work takes maximum advantage of the PI's increasing access to severely septic patients as well as a growing multidisciplinary team that he has assembled for the work.